Fork me on GitHub

Header

Important notice about the «ADG» project

Important notice: Due to large demand, the «Accessibility Developer Guide» project (currently a proof of concept with low priority) will become a top priority priority in 2017. We will make this guide become what it deserves to be for so long already: a comprehensive resource for all folks interested in accessibility. For this we need your support!

How an experienced screenreader user approaches a website

For a sighted person, it's hard to imagine how a screenreader user surfs the web. In this article, we try to give you a concrete idea of the main strategies used by screenreader users when approaching a website.

This article is written from the view of a screenreader user.

Getting a general overview of a website

As a blind person, I don't perceive the full layout of a page intuitively with the first glimpse, the way a sighted person does, so I have to get a general overview of it by combing through its content from top to bottom.

  • When reaching a new website, I press Ctrl + Home to make sure that I'm at the beginning of the document
    • This is important, because sometimes there is some autofocus functionality
    • Autofocus may be helpful, but when I open a site for the first time, I want to read the whole content to have an overview
  • When reaching a new website it's always a surprise which kind of elements are used to structure the content, so I always search for semantical information:
    • Usually I search for headings (by pressing h or 1)
      • Headings should introduce every piece of content, so I assume that this way I find the content content I'm looking for
    • When no headings are present I have to go line by line through the whole content to gain a first impression (using and , not using Tab)
  • I can also get a first impression of the page by using the elements list dialog (NVDA + F7), which (beneath headings and links) shows given landmarks
  • When I encounter a table, then I typically use the table reading commands (Ctrl + Alt + Arrow) so the content of each cell is announced in relation to the corresponding header
  • When I encounter a form and want to interact with it, I press Enter on an input to activate forms mode.
    • From now on, I jump from form element to form element using the Tab key

Interacting with a familiar website

As soon as I know the general structure of a website, I can interact with it much faster.

  • If there are jump links, I use them to jump to the wanted part of the page
  • I can browse directly through certain types of elements like headings (h and 1 to 6), form inputs (keys f and i), links (keys a, v and k), lists (keys i and l), images (g) and so on.
    • For Example, if I open a webmail page, and I want to login, I directly search for an input field by pressing i
    • Or when I open a page on Wikipedia, I press 1 because I know that the first <h1> is the article's heading
  • Sometimes I search for content directly using Ctrl + F